Wednesday, December 10, 2014

Part Two. Responding to the Federal Department of Education's Proposal to Rank Teacher Education Programs: "Teaching, Admissions Standards, and Closing Programs" (Reposted from 1/6/14)

The federal Department of Education is proposing a set of regulations for teacher education programs. There is a 60-day period for the public to comment on these proposals. You can read the Department’s November 25, 2014 press release here.

Some of the proposed regulations are reasonable (graduates giving feedback to their teacher education programs) and some are terribly wrongheaded, repeating the disastrous kind of thinking that shaped No Child Left Behind: for example, teacher ed programs would be evaluated based on the test scores of the children taught by the programs’ graduates. There are logical and conceptual problems as well as technical ones with this proposal, as I argued in Part One of these posts.

Over the next few weeks, I am going to continue to repost the three pieces I wrote in late 2013 and early 2014 on teaching and teacher education. They are relevant to the current discussion of teacher education and to the proposed regulations from the federal Department of Education.

Here is the second.

***

There are many different types of effective teaching and many roads to get there. Travelling across the United States to document good public school classrooms for Possible Lives, I saw solid to extraordinary teachers of many stripes: shy and outgoing; desks in rows and desks all over the place; some were low-key and methodical, and some were energetic and spontaneous; some swore by one way of organizing their curricula and classrooms that others would find unworkable; some spoke a fair amount, others turned the floor continually back over to their students.

            Yet within the variability, there were qualities they all shared. They had command of the material they taught. They created safe and respectful classrooms. They had a deep belief in the ability of their students and held high expectations for them. They required their students to think and think hard and worked to engage students in each others’ thinking. The richness came in the variety of ways they realized these qualities—an important point, given the push by some for increasingly regulated curriculum and pedagogy.

Part of the variation, of course, was a result of where these teachers went to college. But the variation also came from influential teachers they had earlier in their own schooling. The way they taught was also influenced by their personalities and by their values and background: by family or religion or positive or negative experiences in school; by the experience of race or ethnicity, social class, gender, or sexual orientation; by political and social commitments; by the love of a subject. An important quality of a teacher education program, traditional or alternative, is how well it is able to draw on and develop these characteristics. You won’t see this quality mentioned in any of the high-profile reports on teacher education.

With a few exceptions, the teachers in Possible Lives came from modest middle-class to working-class backgrounds. (This tends to be true for teachers as a whole.) A fair number went through local or regional teacher ed programs—the kinds of programs that have been targeted in teacher ed critiques. Because of finances or family expectations or cultural norms, some of the teachers I observed had few other options.

One compelling reason behind the rise of alternative credentialing programs is to draw into the teaching force a wider sweep of people from a range of backgrounds—particularly people who might not otherwise have gone into teaching. This is all to the good. But at the same time there’s this expansive impulse in the discussion and debates around teacher ed, there is also a restrictive counter-force: calls to raise admissions standards into teacher education and recommendations to limit or close particular kinds of teacher ed programs. Let me consider each in turn.

***
Raising Standards

The general complaint here is that traditional teacher education candidates, on average, come from the lower ranks of their class and score below the national average on SAT and GRE exams. There is more variation here than the average suggests, however: some local colleges take many of their applicants, and some universities are quite selective. Also secondary-level candidates tend to have higher grade point averages and test scores than their elementary-level peers. Still it is of course true that we want to do everything possible to draw people with strong educations into teaching. Teaching is intellectual work, as I noted earlier, and I think that one thing that has limited the profession is that teaching—especially at the elementary level—is not typically defined that way. Teachers need a good general education, and, hugely important, I think, they need to be interested in education, gain pleasure from learning and thinking about learning. For those who will teach a particular subject, they need to be well educated in that field.

So, to be clear, knowledge matters. I have seen too many instances of teachers providing superficial or downright incorrect comments on student papers or stumbling through a science or math lesson on material they clearly don’t understand. But knowing something, as fundamental as that is, is half the story; knowing how to teach it—“pedagogical content knowledge,” in psychologist Lee Shulman’s famous phrase—is equally important. We have a history in the United States of defining teaching primarily in terms of process and technique or in terms of subject matter knowledge. Of the many fruitless dichotomies that bedevil education, this is among the most unproductive.

One of my concerns about the contemporary teacher ed debates is that knowledge—as represented by undergraduate major and GPA—is held in some circles as the touchstone of teaching excellence. Certainly a big part of Teach For America’s appeal is the undergraduate pedigree of its interns. And it seems to be the hope of some alternative programs that if we just get more “smart” people—smart defined by academic background, GPA, test scores—into teaching, we will have gone a long way toward solving the “teacher quality” issue. But an undergraduate at our most prestigious colleges and universities can go through four intense years of literature or chemistry and never once be confronted with the question: How would I teach this?

A while back, I spent time doing research in a top-ranked medical school. To a person, the students had through-the-roof academic credentials and did exceedingly well in their first two years of science courses. Talk about smart! The striking thing was that a fair number of them had real difficulty as they moved toward patient care. Not only were they socially inept—distant, awkward—but also diagnostically maladroit, partly because they couldn’t communicate with their patients and partly because of the difference between knowing physiology and using it to diagnose and help cure another human being. In response to this not uncommon state of affairs, medical schools across the country have been modifying supervision; instituting courses in communication, patient care, “doctoring,” and the art of medicine; and changing their recruiting and admissions policies to widen the net, gaining some students who might not have the same astronomical GPAs, but possess other qualities that contribute to being a good doctor.

I think we need to be cautious about conflating academic achievement with the ability to teach. The two are intimately related, but not one and the same.

There is a further issue, and that is the diversity of the teaching force. What happens to our talent pool as we tighten restrictions on who gets into teacher education programs? Who might get left out? Some of the young people who are most passionate about teaching in low-income communities come from those communities, and therefore have probably not had either the in-school or out-of-school resources that contribute to strong post-secondary achievement—particularly for certain majors. This scenario does not hold true for all students coming out of low-income schools, but for enough to concern us here. Passion alone does not warrant entry into the teaching profession by any means. If our candidates still need to further develop their academic knowledge and skills in certain areas, then they must do so before or during their teacher ed program—and the program needs to hold them accountable. But to systematically exclude them in a country so beset by structural inequalities is to bar from the classroom a group of people most familiar with the barriers low-income students face and deeply committed to helping those students get a better education than they did.

One last point. Another argument in the air for raising admissions is that a higher entrance bar will enhance the status of the teaching profession. Countries such as Finland are invoked where teachers face tough entrance criteria and enjoy solid professional status. These kinds of claims, and the invoking of other countries to support them, reveal one of the problems in the teacher ed debates: a tendency to make simplified causal connections and cross-cultural comparisons. Reading the sociological scholarship on the development of professions reveals what a complex process professionalization is—influenced by cultural traditions, politics and economics, gender and racial dynamics, the role of advocacy organizations and powerful leaders, and more. And the way these factors played out for teaching over the last century in the United States and Finland are pretty different.

Raising teacher ed entrance requirements in our country might have some effect on occupational prestige, but it would be one of many factors determining status, more potent ones being salary, gender bias, and degree of occupational autonomy. There may well be good reasons for a particular teacher education program or group of programs to raise its admissions standards, but that decision would need to be made after careful analysis of potential benefits and liabilities for its region and not on simplistic sociological abstractions.

***
Closing Teacher Education Programs

            Though programs in all types of colleges and universities come in for criticism in the major teacher education reports, those housed in less prestigious institutions that produce over fifty percent of our teachers—regional state universities, small public and non-selective private colleges—take a hard hit.

Consider two reports that got a good deal of media attention: Arthur Levine’s 2005Educating School Teachers (mentioned in my first post) and Teacher Prep Review, released by the National Council on Teacher Quality (NCTQ). Though both reports single out a (quite small) number of good programs—and those programs range in type and size—the overall assessment they present is devastating.

The Levine report recommends closing poor programs, many of which, he believes, are in those state universities and non-selective colleges. In turn, the programs that should be expanded are located in research universities. (This same advocacy for research university programs runs through other, earlier reports, such as that by the 1980s Holmes Group.)

The National Council on Teacher Quality was able to rate 1,200 programs, and it placed about 15% of them so low as to warrant a “consumer alert.” Many of the 15 percent, though not all, are the same kinds of programs Levine criticizes. The authors of the NCTQ report hope that the warning will lead prospective teachers to vote with their feet and go to other schools (and school administrators to look elsewhere for new hires), thus forcing the targeted schools to improve or go out of business.

It is not at all my purpose here to defend poor programs, or even to dispute the possibility that, on average, sub-par programs might be found more in one category of institution than another. But I do want to raise several concerns.

There’s an assumption in some of the reports—clearly stated in the one from NCTQ—that students interested in a teaching career are free agents, able to make the classical economists’ rational choice about benefits and losses, and act accordingly. They are able to go to the school that will provide the greatest payoff. But, as I noted earlier, some students are not in a financial or personal position to make such a choice. The local teacher ed program is their only option. Reading these reports, one gets the sense that the authors are at a great social distance from the lives of such students.

Some of the reports also operate at a real distance from the colleges and universities they criticize. What struck me about several of the small out-of-the-way programs I visited during my travel for Possible Lives was how embedded they were in their communities, how well the faculty understood the kids in the schools, the local history, the social and economic pressures on the region. Some of the faculty themselves went to local, non-elite colleges or universities, they didn’t publish in scholarly journals, they didn’t have the bonafides of their contemporaries in snazzier institutions. But they were smart and skillful, and they provided substantial support to the novice teachers in their charge: mentoring them, meeting with them after hours, observing them teach.

These were two good programs, and I bring them up not to generalize from them, but to illustrate a point about analysis from a distance. The National Council on Teacher Quality report could not get to the qualities I sketched. It is built primarily on analysis of course descriptions and syllabi. These will provide course philosophy and purpose in an abbreviated form, reading lists and topics, assignments, grading criteria, and the like. Little more. The Levine report utilized a much more comprehensive methodology: surveys of principals and education school administrators, faculty, and alumni; site visits to twenty-eight ed schools; and statistical analysis of data on program graduates and the students in the schools where they teach. The site visits focused on institutional structure, governance, and demographics, but I was not able to tell from the report if the visits also got to the more experiential level that I raise here.

The Levine report, as substantial as it was, raised other concerns about policy recommendations for categories of institutions. Let me provide one example, for it represents a kind of reasoning we see all too often in current education debates.

Levine commissioned a study to compare the reading and math scores of students by the type of teacher ed program their teachers attended. The statistically significant results demonstrated that students who were taught by teachers who attended research universities showed one-and-one-half weeks more growth in math than students taught by teachers who attended the aforementioned less-selective institutions. “Over the course of 12 years of schooling,” Levine writes, “this amounts to four and a half months” of growth. A result like this gets shortened in debate and opinion pieces to damning evidence that a whole slew of teacher ed programs produce poorly trained teachers. Let’s consider this result, and the reasoning that leads from it to a significant policy recommendation.

It’s important to remember that, though ambitious, this is a single study that would need to be replicated. Furthermore, the difference of one-and-one-half weeks of growth over a school year is in fact a small difference or “effect size,” and it gains statistical significance because of the large numbers of students and teachers in the sample. Effect size is a basic issue in such analyses: one can run a technically flawless analysis with a large sample size and get a statistically significant result, but the important question is whether that result matters enough to lead to a decision to act—in this instance to build a case for closing or scaling back a group of teacher ed programs.

There is also a logical problem at the heart of this example. Levine extrapolates from a single one-year study and projects out over 12 years. (Let’s put aside for a moment my contention that the effect size here is not alarming—nor, following Levine, is four-and-one-half months over a 12 year period.) For the score differential found in one year to maintain itself over 12 years requires that all other factors in the lives of the children and their schools remain the same: that the students maintain the same level of motivation, don’t get sick, don’t experience family disruption. That teachers are equally immune from life’s perturbations, and when that is not the case, they are quickly replaced. That the school-level leadership doesn’t change; that new policies aren’t enacted; that funding remains stable; that the community isn’t hit with economic hardship; and so on. The 12-year extrapolation assumes an “other things being equal” statistical model in a world where very little remains equal. Such extrapolations make for dramatic statements, but they are not conceptually sound and should not be part of the logic of a policy recommendation that would have serious consequences for many regions of the country.


As with any institution, there is a range of quality in teacher education programs, and some are beset by the kinds of problems Levine and others identify: poor leadership, a fragmented curriculum, inadequate opportunity for students to engage in classroom practice. If in fact more troubled programs exist in the category of institution that produces over one-half of our teachers, then one would think that an important educational and social agenda would be to focus on ways to help them improve where possible and not to advocate for their abandonment and closure.

You can share this blog post on Facebook, Twitter, or Google Reader through the "share" function located at the top left-hand corner of the blog. 

No comments:

Post a Comment